
Ryan Morris
157 posts

Ryan Morris
@ryanmorris59
Building companies that will change the world, end pollution, empower individuals




Howard Marks: "When you buy the S&P 500 at a 23x P/E, your 10-yr annualized return has always fallen between +2% and –2%, IN EVERY CASE, EVERY CASE!"


Simply adding Gaussian noise to LLMs (one step—no iterations, no learning rate, no gradients) and ensembling them can achieve performance comparable to or even better than standard GRPO/PPO on math reasoning, coding, writing, and chemistry tasks. We call this algorithm RandOpt. To verify that this is not limited to specific models, we tested it on Qwen, Llama, OLMo3, and VLMs. What's behind this? We find that in the Gaussian search neighborhood around pretrained LLMs, diverse task experts are densely distributed — a regime we term Neural Thickets. Paper: arxiv.org/pdf/2603.12228 Code: github.com/sunrainyg/Rand… Website: thickets.mit.edu


Octopuses on MDMA start hugging each other. They're solitary animals that normally avoid all physical contact. One molecule turned a creature with three hearts into a cuddler. Now imagine what it does for a species that was built for connection but forgot how.






















